cover peter mombaerts

Peter Mombaerts on olfactory system and odorant receptor genes

  • cover play_arrow

    PLAY EPISODE


cover peter mombaerts
Season 2013
Season 2013
Description arrow_drop_down

Description

How does a mouse nose with 1,200 receptor genes wire itself into a precise sensory map, and why is that map less stereotyped than we once believed? Peter Mombaerts explores the genetics and development of olfactory circuit formation.

Subscribe for more from the Convergent Science Network podcast series.

Peter Mombaerts describes the remarkable complexity of the mouse olfactory system, where approximately 1,200 odorant receptor genes each expressed by a distinct population of sensory neurons must organize their axonal projections into roughly 3,600 glomeruli in the olfactory bulb. Using gene targeting and molecular labeling techniques, his laboratory has spent two decades investigating how this glomerular map develops and what role the receptor proteins themselves play in axon guidance and glomerular identity.

A central theme is the surprising degree of variability in glomerular positioning. Mombaerts challenges the widely used term “stereotyped” to describe the glomerular map, demonstrating that even between the left and right bulbs of the same inbred mouse, the relative positions of identified glomeruli can be inverted. He prefers terms like “recognizable” or “reproducible,” noting that the precision is insufficient to construct a definitive atlas as has been done for Drosophila. This variability has important implications for understanding the mechanisms of map formation: if the map were truly stereotyped, extremely complex molecular guidance mechanisms would be required, but acknowledging the jitter relaxes these demands considerably.

The conversation explores the genomic organization of odorant receptor genes, which Mombaerts describes as “haphazard,” distributed across approximately 40 loci with the largest cluster containing around 300 genes. Expression levels vary over two orders of magnitude between different receptor types, and the one-neuron-one-receptor rule, while strongly supported, remains an asymptotic conclusion. Remarkably, replacing an odorant receptor’s coding region with the beta-2 adrenergic receptor still produces neurons that form a recognizable glomerulus and respond to appropriate ligands, suggesting that the receptor protein’s role in axon guidance may not be unique to olfactory receptors.

The episode also addresses the emerging recognition that odorant receptor genes are expressed outside the nose, including in kidneys where knockout of one receptor affects blood pressure regulation, hinting at broader biological roles for this gene family beyond olfaction.

Tagged as:

About the author call_made

CSN Podcasts

Both the triumphs of humanity and its most evil deeds have resulted from collaboration. In a time where humanity is required to aspire to the former and minimize the latter, the question arises of how collaboration arises and why it fails. Surprisingly, this phenomenon, so central to who we are, is not well understood. Hence, a collaborative effort is required to understand collaboration in its full biological, psychological, sociological, cultural, and economic complexity and to translate this understanding into operational impact. This series of podcasts is one step toward achieving these complementary goals. The Collaboration Podcast presents interviews with people who are central orchestrators of collaboration in various domains including business, government, science, art, health, sustainability, and the military. The discussions were conducted by Prof. Dr. Paul F.M.J. Verschure and members of the Program Advisory Committee of the Ernst Strungmann Forum on Collaboration (https://www.esforum.de/forums/ESF32_Collaboration.html) during 2021 and had the goal to sketch a map of opportunities, challenges, and obstacles in human collaboration. The forum took place in May 2022, and now we would like to share this series of interviews with a broader audience. The full report of the Forum will be published in 2023 by MIT Press. The podcast was produced by the Convergent Science Network (https://www.convergentsciencenetwork.org/). Context: The stability of social systems depends critically on realizing sustainable methods of “collaboration,” yet how and by which means collaboration is achieved is not clearly understood; neither are the conditions or processes that lead to its breakdown or failure. Collaboration can be understood as cooperation between agents toward mutually constructed goals. Part of the reason for our lack of understanding is that the phenomenon of collaboration is, by nature, a highly multidisciplinary problem, and effective research into its complexities has been difficult to achieve across the broad range of scientific and technical disciplines involved. The need for a fundamental understanding of collaboration, however, has become increasingly important. Not only does humankind demand answers as it attempts to address critical challenges at multiple scales (e.g., climate change, migration, enhanced automation, social and economic inequality), but ever-increasing technological and economic means of interconnecting people and societies are disrupting long-established, familiar patterns of how we interact. Radical technological changes that are ongoing have the potential to reshape collaboration in ways that are currently hard to predict or influence (e.g., by altering configurations in interaction, information creation, and modes of communication). On one hand, such changes could disrupt hitherto stable forms of collaboration by affecting critical communication channels and traditional roles, as can be observed in the rapidly changing patterns in governance, commerce, and social interaction. Conversely, technology could lead to the emergence of novel, successful forms of collaboration that deviate from traditional “hierarchical” architectures. Evidence of this can be seen in areas as diverse as highly automated manufacturing plants, the open science movement, collaborative software repositories, user-centered services, and the sharing of economy-based modes of organization. Without a fundamental understanding of the mechanisms, processes, and boundary conditions of collaboration, it is not possible to evaluate or predict which of these possible scenarios are sustainable or even plausible. The Forum “How Collaboration Arises and Why it Fails” (May 8–13, 2022, Location: Frankfurt am Main, Germany) Chairs: Andreas Roepstorff and Paul Verschure Program Advisory Committee: Jenna Bednar, Julia R. Lupp, Bhavani R. Rao , Andreas Roepstorff, Ferdinand von Siemens, and Paul Verschure

More posts

Timestamp

  • fast_forward00:00:03 - This is the Convergent Science Network podcast. Leading researchers in the domain
  • fast_forward00:00:10 - of neuroscience, brain theory and technology are interviewed by Paul Verschure and Tony Prescott.
  • fast_forward00:00:26 - This is Paul Verschure with the Convergent Science Network. I'm here with Peter Mombarts.
  • fast_forward00:00:33 - And Peter is a neuroscientist who has been specialized in the development of
  • fast_forward00:00:38 - the nervous system, the wiring of nervous systems.
  • fast_forward00:00:40 - And Peter, you were telling us this morning that you chose the olfactory system
  • fast_forward00:00:48 - of the mouse as your target system. So why is that?
  • fast_forward00:00:53 - At heart, I'm a geneticist, molecular biologist. And when these auto-receptor
  • fast_forward00:00:59 - genes were discovered in 1991,
  • fast_forward00:01:00 - I felt that there was a great way of approaching the nervous system,
  • fast_forward00:01:08 - particularly development, in a genetic and molecular way because that's what I'm at heart.
  • fast_forward00:01:13 - And indeed, with time, we have learned that using these organ receptor genes,
  • fast_forward00:01:21 - of which there are a very large number in the mouse, 1,200,
  • fast_forward00:01:24 - we can differentially stain and manipulate populations of olfactory neurons,
  • fast_forward00:01:30 - each expressing one of these receptors.
  • fast_forward00:01:32 - And that's a very unique experimental advantage that the olfactory system of
  • fast_forward00:01:36 - the mouse offers at this time. Right.
  • fast_forward00:01:38 - So that was really a big discovery in the early 90s, right? That the idea that
  • fast_forward00:01:43 - actually individual receptor neurons would be, if you want, tagged by single
  • fast_forward00:01:48 - genes or would express single genes.
  • fast_forward00:01:52 - Yes, the history of research in olfaction, in my view, can be divided in a pre-
  • fast_forward00:01:58 - and post-1991 era, with the paper of Linda Bock and Richard Axel,
  • fast_forward00:02:04 - published in April 1991.
  • fast_forward00:02:07 - One, and it was the discovery of these auto-receptor genes that really made
  • fast_forward00:02:12 - a proper experimental investigation of this system possible.
  • fast_forward00:02:18 - At that time, it was not known that there would be only one gene expressed per neuron.
  • fast_forward00:02:23 - It looked like that from the beginning, that a small number of genes would be
  • fast_forward00:02:27 - expressed by a given cell, and more and more the evidence is consistent,
  • fast_forward00:02:32 - perhaps asymptotically, with this one neuron, one gene rule.
  • fast_forward00:02:37 - So that in some sense, now given that you were exposed to this discovery,
  • fast_forward00:02:44 - you saw this as an opportunity to now have a specific preparation to understand
  • fast_forward00:02:49 - how this system might wire itself.
  • fast_forward00:02:52 - But why in particular an olfaction? You could also have gone for,
  • fast_forward00:02:55 - let's say, any other system where you might have some genetic label to look at wiring.
  • fast_forward00:02:59 - So what makes the olfactory system so appropriately sculpted for that?
  • fast_forward00:03:04 - Well, not all decisions made by scientists are rational.
  • fast_forward00:03:07 - There's also something called gut feeling. And back then I felt this was a wonderful new field.
  • fast_forward00:03:13 - Also, it was new, right? It was a new approach.
  • fast_forward00:03:16 - And I haven't regretted since then. There may be more interesting systems for
  • fast_forward00:03:20 - other questions. But the questions I started to get interested in about 20 years
  • fast_forward00:03:25 - ago are still the questions I'm
  • fast_forward00:03:28 - interested in now that I think about every morning when I take a shower.
  • fast_forward00:03:31 - And as long as that's the case, I think I will continue to work on it.
  • fast_forward00:03:35 - So now, what's the – can you give – so this morning, you spent quite some time
  • fast_forward00:03:40 - to sketch out for us the basic structure of this mouse olfactory system.
  • fast_forward00:03:45 - So before we start looking at the details of your discoveries,
  • fast_forward00:03:49 - what are the key elements of this olfactory system that we should keep in mind?
  • fast_forward00:03:55 - So anatomically, a molecular, there is a division in different chemosensory
  • fast_forward00:04:01 - structures in the nasal cavity, in the nose.
  • fast_forward00:04:04 - You have the main olfactory epithelium, which contains several million olfactory sensory neurons.
  • fast_forward00:04:10 - We don't really know the number, by the way, but it's several million,
  • fast_forward00:04:13 - each expressing expressing one of these 1,200 auto-receptor genes that Bocanaxo described.
  • fast_forward00:04:20 - Then you have the septal organ, a specialized structure, which contains approximately 10,000,
  • fast_forward00:04:26 - olfactory neurons, and half of these express the same receptor called SR1,
  • fast_forward00:04:30 - which is a rule breaker as these neurons are activated by a very wide variety of chemicals.
  • fast_forward00:04:37 - We have the vomeronasal system, abbreviated VNO, vomeronasal organ,
  • fast_forward00:04:43 - which doesn't express these auto-receptor
  • fast_forward00:04:45 - genes, but two other families of G-protein-coupled receptor genes,
  • fast_forward00:04:50 - V1R genes and V2R genes, V-vomeronasal, approximately 300 in total.
  • fast_forward00:04:55 - And then finally, a very small chemosensory structure at the tip of the nose
  • fast_forward00:05:00 - called the The Grüneberg ganglion, abbreviated GG, maybe 500,000 cells,
  • fast_forward00:05:05 - that's perhaps mostly active in neonatal newborn mice.
  • fast_forward00:05:11 - So there's already quite some complexity of chemosensory anatomical structures
  • fast_forward00:05:15 - in the nose and the corresponding repertoires of chemosensory receptor genes.
  • fast_forward00:05:20 - Not surprisingly, as mice are highly chemosensory organisms,
  • fast_forward00:05:27 - they rely less on their sense of vision as we do, for instance.
  • fast_forward00:05:33 - These neurons are all projecting one axon, one nerve ending,
  • fast_forward00:05:38 - to the brain, to the main olfactory bulb or the accessory olfactory bulb.
  • fast_forward00:05:44 - And in the bulb, each axon terminates only in one of the so-called glomeli,
  • fast_forward00:05:52 - of which are about 3,600 in the mouse.
  • fast_forward00:05:56 - So for the main olfactory system, which detects what we call garden-variety
  • fast_forward00:06:00 - type of odorants, The picture is as follows. We have 1,200 genes.
  • fast_forward00:06:06 - Most of these are expressed in olfactory neurons, but only one gene per cell, per neuron.
  • fast_forward00:06:12 - Each neuron has one axon, and it enters one glomulus in the bulb,
  • fast_forward00:06:17 - where it then synapses with second-order neurons.
  • fast_forward00:06:21 - Okay. So now we have this structure, which is fairly layered.
  • fast_forward00:06:26 - And so it seems to have it must have some tight control over how it is this
  • fast_forward00:06:32 - is wired up and I think an added complexity of the system is that actually there's
  • fast_forward00:06:37 - a continuous turnover of these receptor neurons so now the real problem becomes okay,
  • fast_forward00:06:44 - this layer where it hits the olfactory bulb these nerve endings form then these
  • fast_forward00:06:48 - glomeruli and how is the specificity of the organization of this glomeruli now assured.
  • fast_forward00:06:59 - So during development, and quite quickly during development,
  • fast_forward00:07:02 - I would say the first few days after birth, this glomerular array matures,
  • fast_forward00:07:08 - finally forming these approximately 3,600 glomeruli, and the neurons that express
  • fast_forward00:07:14 - the same receptor project their axons to the same glomeruli.
  • fast_forward00:07:19 - In fact, they form these glomeruli. It's not that the glomeruli are there and
  • fast_forward00:07:22 - they just have to project their accents to them.
  • fast_forward00:07:25 - And this glomalor array, or glomalor map if you wish, develops quite reproducibly,
  • fast_forward00:07:31 - left and right bollop in the same individual and from mouse to mouse.
  • fast_forward00:07:35 - It's not so reproducible that it's stereotyped, we cannot draw a map with coordinates
  • fast_forward00:07:40 - on the bollop and say this is this glomalus for this receptor,
  • fast_forward00:07:43 - but it's very reproducible, or recognizable.
  • fast_forward00:07:49 - What that means for the function of function for decoding is not clear,
  • fast_forward00:07:55 - but there is clearly a spatial map that develops quite early,
  • fast_forward00:07:59 - very precisely in the mouse.
  • fast_forward00:08:02 - And also you then went to describe how this kind of labeling could occur.
  • fast_forward00:08:09 - So there was this idea of zones, right? Maybe there are zones that sort of define
  • fast_forward00:08:12 - a sort of a rough kind of chemotopic map in which you can then place your glomeruli. In the olfactory bulb.
  • fast_forward00:08:19 - In the olfactory bulb, yes. Do you think that's still a plausible scheme?
  • fast_forward00:08:22 - So there have been several attempts and several papers over the years in different
  • fast_forward00:08:26 - model systems to ask with various electrophysiological or other methods,
  • fast_forward00:08:34 - are there particular types of chemicals that certain regions of the bulb are
  • fast_forward00:08:39 - more sensitive to than others?
  • fast_forward00:08:42 - That's called chemotopy. entropy so there would be a chemical structure somehow
  • fast_forward00:08:46 - to the glomerular array we don't know of course a priori what are the valid
  • fast_forward00:08:50 - features that would be extracted is it chain length.
  • fast_forward00:08:56 - Saturated unsaturated bonds molecular weight and so on we don't really know
  • fast_forward00:08:59 - this in advance but have been attempts and there is some evidence that finds
  • fast_forward00:09:03 - that and some other evidence that doesn't find it to some extent one can find
  • fast_forward00:09:07 - what one wants to find right and no but wait But there's an important thing you said earlier.
  • fast_forward00:09:12 - You said, okay, across individual mice, you find a rough chemotopic map.
  • fast_forward00:09:18 - There's just some jitter with this glomerular, how it's placed in that, if you want, matrix.
  • fast_forward00:09:23 - Yeah. But that would suggest that there is some sort of zoning,
  • fast_forward00:09:28 - if you want, of this bulb in development.
  • fast_forward00:09:33 - So the chemical logic of the glomerular array is not overwhelming.
  • fast_forward00:09:39 - It's not the case that each part of the bulb is equally sensitive to all odorants,
  • fast_forward00:09:43 - right? That would be the other extreme.
  • fast_forward00:09:44 - Yes, there is some kind of regionalization, but it's not overwhelming and we
  • fast_forward00:09:48 - don't really also know in the end what that means functionally,
  • fast_forward00:09:51 - you know, if it's a byproduct of it doesn't make sense.
  • fast_forward00:09:55 - No, let's do a thought experiment. I mean, if we take a mouse,
  • fast_forward00:09:58 - we present it with lemon.
  • fast_forward00:10:01 - Lemon, now we see some glomerulus light up in the olfactory bulb,
  • fast_forward00:10:06 - and I ask you, go show, go find that same glomerulus now in another mouse.
  • fast_forward00:10:11 - Then you would take the location of this identified glomerulus to start your search.
  • fast_forward00:10:17 - What's the probability distribution around that point in space to find the glomerulus
  • fast_forward00:10:22 - that will also respond to this lemon odor?
  • fast_forward00:10:29 - There have been a few studies looking at that anatomically, and the level of
  • fast_forward00:10:32 - uncertainty, if you wish, of variability is maybe in the order of 1.52%.
  • fast_forward00:10:38 - So even between the left and the right ballop, that's probably a better way of looking at it.
  • fast_forward00:10:41 - If you know the positions of glomeruli on the left ballop, you can't really,
  • fast_forward00:10:45 - to the level that we would like to, find the positions of the glomeruli on the
  • fast_forward00:10:50 - other ballop. We cannot make an atlas, in other words.
  • fast_forward00:10:52 - Unlike, for instance, in the honeybee or the drosophila, where people have been
  • fast_forward00:10:57 - able to make an atlas and say precisely this glomulus and this position with
  • fast_forward00:11:01 - this shape, volume, and so on, that is that glomulus, and they give it a name.
  • fast_forward00:11:05 - That we cannot do with the mouse or the rat thus far.
  • fast_forward00:11:10 - But implicitly, I have the feeling you're saying that you believe that there's
  • fast_forward00:11:15 - not a strong structuring.
  • fast_forward00:11:16 - But in some sense, if we would try to do this more quantitatively and say,
  • fast_forward00:11:20 - okay, if I have X, Y, Z of an identified glomerulus, now I need X,
  • fast_forward00:11:25 - Y, Z plus some delta to find in another animal with probability 90%, let's say.
  • fast_forward00:11:30 - Yeah, there's certainly a… How big is that delta? Yeah, there's certainly a probability.
  • fast_forward00:11:33 - So I think the best way of visualizing this variability.
  • fast_forward00:11:39 - Is to look at two glomeruli at the same time, for instance, of two strains of
  • fast_forward00:11:44 - mice with a different marker, maybe one like Z, the other one GFP,
  • fast_forward00:11:48 - cross them together, And then glomalus A and B would be on the left side, for instance,
  • fast_forward00:11:54 - A more lateral than B, and on the right side, it may be the other way around.
  • fast_forward00:12:00 - And this is more than whole distortions of the map.
  • fast_forward00:12:05 - You could think perhaps that the bulb would be a little bit squeezed and things
  • fast_forward00:12:08 - would be a little bit more anterior, posterior.
  • fast_forward00:12:09 - Here, but if you have internal rearrangements, medial-lateral relative positioning
  • fast_forward00:12:14 - that is inverted, then that clearly shows there is an underlying variability,
  • fast_forward00:12:18 - which we have to acknowledge, unfortunately, to the point that we cannot make,
  • fast_forward00:12:22 - even in inbred mice, a real atlas.
  • fast_forward00:12:27 - And one actually may then also wonder if there is really a glomelon map.
  • fast_forward00:12:31 - That word gets used a little bit loosely.
  • fast_forward00:12:34 - Depends what kind of resolution and detail you want, but one can actually wonder,
  • fast_forward00:12:38 - is Is that really a glomerular map for a species?
  • fast_forward00:12:40 - But now if I would take my two mice again and we have our identified glomerulus,
  • fast_forward00:12:47 - how big is the probability that I would find the matching glomerulus in another
  • fast_forward00:12:52 - animal in exactly the opposite position? Yeah.
  • fast_forward00:12:57 - Yeah, that I don't know. Maybe we have only emphasized the cases where there is a mismatch. Okay.
  • fast_forward00:13:01 - But if you look more like the circle or the area, I would say about 20 or 30
  • fast_forward00:13:06 - glomeruli for the two or three studies where this has been done.
  • fast_forward00:13:09 - So 20 or 30 glomeruli out of 3,600 is quite small, actually,
  • fast_forward00:13:14 - if you wish. That's what I'm saying.
  • fast_forward00:13:15 - Yeah. So you could imagine that you have a rough patterning with a resolution of 10 to 20 glomeruli.
  • fast_forward00:13:20 - And that's still, what, 10% of your total structure?
  • fast_forward00:13:25 - 1%. Oh, 1%, exactly 1%. And then within that, okay, you have some jitter.
  • fast_forward00:13:31 - But this is actually pretty precise, I would say, as sort of a non-expert in this domain.
  • fast_forward00:13:37 - Well, from a practical point of view, it's not precise enough for us to make
  • fast_forward00:13:40 - some kind of an atlas, right?
  • fast_forward00:13:42 - Which we thought a few years ago we could do. We could make a probabilistic atlas. Exactly right.
  • fast_forward00:13:48 - But maybe we, in Drosophila again, although the number of glomeli and receptors is much lower,
  • fast_forward00:13:56 - there you can there it's really been possible to do so yeah but the interesting
  • fast_forward00:14:00 - thing is of course you apparently were expecting to find some precision in this
  • fast_forward00:14:04 - atlas that was not satisfied no i i wasn't but but others were and they and
  • fast_forward00:14:08 - they use terms even now still in print of,
  • fast_forward00:14:11 - stereotyped which from the old french type i think the stereotype where you
  • fast_forward00:14:16 - really make a photocopy right you have some kind of a template and you print
  • fast_forward00:14:19 - an exact copy with maybe some little bit of dirt or so but it's basically an
  • fast_forward00:14:23 - exact copy and that's not really the case and i wouldn't use that term either.
  • fast_forward00:14:26 - What we like to say after thinking about this for quite a while,
  • fast_forward00:14:30 - recognizable or reproducible, something like that.
  • fast_forward00:14:33 - It's not stereotyped. You cannot even from the position of the glomalin,
  • fast_forward00:14:37 - the left ballop predicts.
  • fast_forward00:14:39 - With high degree of certainty, those in the right bulb. Okay.
  • fast_forward00:14:43 - But then would you still go along with the idea that it's a rough patterning
  • fast_forward00:14:47 - and then through some self-organizing process, this gets filled in and out of
  • fast_forward00:14:53 - result of the self-organization of the variability?
  • fast_forward00:14:55 - Or would you say at this stage, look, it's better not to commit ourselves too
  • fast_forward00:14:59 - strongly to the idea of a map because it's too ambiguous at this stage?
  • fast_forward00:15:03 - Yeah, and it can also be misleading to think in terms of mechanisms, right?
  • fast_forward00:15:09 - If you really believe, I think some people still believe that this map is extremely
  • fast_forward00:15:14 - precise and is in variable positions, then the mechanisms that enable to produce
  • fast_forward00:15:21 - that must be extremely complicated, right?
  • fast_forward00:15:24 - But if you acknowledge that there is indeed a bit of jitter or variability or
  • fast_forward00:15:28 - uncertainty, then it relaxes a little bit the demands you would put on these mechanisms.
  • fast_forward00:15:34 - So it's more than just being a little bit too strict or philosophical.
  • fast_forward00:15:39 - We really want to know what kind of precision do we expect from the mechanisms
  • fast_forward00:15:44 - that put that into place, which are largely unknown in my view today. Right.
  • fast_forward00:15:50 - But now on top of that, it's not a case that we can think about this glomeruli
  • fast_forward00:15:54 - as being really clean sphere type structures that are just neatly stacked together.
  • fast_forward00:16:00 - You also have quite some variability in the shape of this glomeruli,
  • fast_forward00:16:03 - so this cluster of nerve endings themselves.
  • fast_forward00:16:06 - So what kind of variability do you see there? So yeah, glomalus is acellular.
  • fast_forward00:16:12 - It doesn't have any cells in it, no nuclei.
  • fast_forward00:16:14 - It's a few.
  • fast_forward00:16:18 - Hundreds of incoming axons that terminally arborize and make synaptic contact
  • fast_forward00:16:22 - with the dendrites of the second-order neurons.
  • fast_forward00:16:25 - So this ball of synapses, that is a glomelus, which is a discrete structure
  • fast_forward00:16:30 - surrounded by some glia and pyroglomelar neurons and so on.
  • fast_forward00:16:35 - We would like them to be, of course, all perfect spheres, but that's indeed
  • fast_forward00:16:39 - not the case. Some are a bit more oblongated.
  • fast_forward00:16:43 - Some are in the form of a haltier with a central stalk and two balls hanging
  • fast_forward00:16:47 - on it, which makes it a bit difficult sometimes to say is this one or two glomeruli.
  • fast_forward00:16:52 - The average diameter would be about 55 micron in the Mars, but also there,
  • fast_forward00:16:56 - there is quite some difference.
  • fast_forward00:16:59 - So it's true, it's not 3,600 perfect spheres that are aligned.
  • fast_forward00:17:04 - Moreover, there are different layers in the dorsal part of the bollop.
  • fast_forward00:17:10 - It's typically only one glomerulus, but
  • fast_forward00:17:13 - more ventrally in the ball up there you have multiple glomeruli on top of each
  • fast_forward00:17:17 - other and deeper into the tissue and so on so there is an another another dimension
  • fast_forward00:17:22 - there right so then in but whatever the the structuring that takes place to
  • fast_forward00:17:29 - get the patterning of the glomeruli,
  • fast_forward00:17:31 - there is a belief that that the the single genes expressed by the receptor neurons
  • fast_forward00:17:37 - that in the end would also give
  • fast_forward00:17:38 - identity to their exons, would in some way play a role in this process.
  • fast_forward00:17:43 - Yes, that's been known for quite a while from genetic data. You can replace
  • fast_forward00:17:48 - the coding region of a receptor by that of another one.
  • fast_forward00:17:52 - In fact, we have even done in one case with an other G-protein copper receptor,
  • fast_forward00:17:56 - the beta-2-anergic receptor, and you get a glomalus.
  • fast_forward00:18:00 - Again, a reproducible, recognizable place.
  • fast_forward00:18:04 - In some cases, it looks like it's a completely new glomalus that wasn't there
  • fast_forward00:18:08 - before, but there is of course space if you have 3600 for a few more.
  • fast_forward00:18:13 - You can also make smaller mutations down to point mutations in other receptors
  • fast_forward00:18:19 - using gene targeting and you can create novel glomelar identity,
  • fast_forward00:18:26 - so glomelar that probably are not there before.
  • fast_forward00:18:28 - The receptor protein is also there along the axons all the way to the end and early in development.
  • fast_forward00:18:36 - Suggesting that it has something to do mechanistically with fasciculation,
  • fast_forward00:18:40 - coalescence, convergence.
  • fast_forward00:18:42 - Exactly how that works at the molecular level is, I don't think,
  • fast_forward00:18:47 - clear, or at least there is no consensus of how an old receptor could do that
  • fast_forward00:18:51 - with such precision again.
  • fast_forward00:18:53 - So then you spent quite some time then to try to show or identify the location
  • fast_forward00:19:01 - of these different genes that the receptor neurons express in the genome.
  • fast_forward00:19:07 - So what kind of pattern do you find there?
  • fast_forward00:19:11 - The pattern I like to, the term I like to use is actually haphazard,
  • fast_forward00:19:16 - which means maybe there is no clear pattern, but maybe it's not excluded that there is a pattern.
  • fast_forward00:19:21 - There are in the mouse approximately 40 loci sites in the genome where you can
  • fast_forward00:19:26 - find old receptor genes.
  • fast_forward00:19:28 - Eight of them are solitary genes, which means megabases up and downstream,
  • fast_forward00:19:35 - there is no other receptor gene in the genome.
  • fast_forward00:19:38 - These are the exceptions. Most genes come in clusters.
  • fast_forward00:19:41 - The largest cluster is approximately 300 auto-receptor genes,
  • fast_forward00:19:44 - all stacked one next to the other.
  • fast_forward00:19:46 - And the typical cluster has no other genes in them, non-auto-receptor genes,
  • fast_forward00:19:51 - although of course there are also exceptions.
  • fast_forward00:19:54 - So this pattern, this distribution is, I would like to say, haphazard.
  • fast_forward00:20:00 - That there is no clear logic to it.
  • fast_forward00:20:02 - It's not that every chromosome has one cluster or there are three huge ones
  • fast_forward00:20:06 - or each cluster of genes is expressed in the same zone in the epithelium or
  • fast_forward00:20:10 - encodes genes of the same family. Even that is not the case.
  • fast_forward00:20:14 - For lack of a better term, and I welcome new terms, I'd like to call it haphazard.
  • fast_forward00:20:19 - But how many genes do we have in the mouse?
  • fast_forward00:20:23 - Genes with an intact open reading frame, approximately 1,200.
  • fast_forward00:20:27 - In fact, the mouse genome is not completely sequenced. We check our favorite
  • fast_forward00:20:31 - clusters once in a while, and the number of genes decreases,
  • fast_forward00:20:35 - actually, from month to month.
  • fast_forward00:20:36 - There are genes disappearing that are probably mistakes of assembly.
  • fast_forward00:20:39 - That's a recent experience. So the jury is still out, but the number seems to
  • fast_forward00:20:44 - converge to 1,200, maybe 1,100.
  • fast_forward00:20:47 - But the total mouse genome?
  • fast_forward00:20:49 - Probably 20,000, 25,000, so quite a significant fraction, yeah. Right, exactly.
  • fast_forward00:20:54 - So here we have these 20,000 genes, but now of those 1,200 that relate to the
  • fast_forward00:21:02 - receptor neurons, how much is the variability in base pairs of these genes?
  • fast_forward00:21:07 - If you take two random odontoreceptor genes, the average is a nucleotide amino acid.
  • fast_forward00:21:18 - I think amino acid homology was 33% or something like that. It's not very high.
  • fast_forward00:21:23 - There are some motifs in the amino acid sequence.
  • fast_forward00:21:27 - One of them is May-Dry-Vac. Another one is Fast-Cash. Easy to remember.
  • fast_forward00:21:32 - Those are kind of typical for autumn receptors.
  • fast_forward00:21:36 - If you see those two in a sequence, you have a very, very high likelihood that
  • fast_forward00:21:40 - these are autumn receptors.
  • fast_forward00:21:43 - So they come in families. You can define a family however you want.
  • fast_forward00:21:47 - There have been, of course, debates about that. If it's 80% cut-off,
  • fast_forward00:21:50 - 60% cut-off. a nomenclature developed in 2002 by Stuart Firestone was based
  • fast_forward00:21:55 - on these families and he was in the 200-something families of these 1,200 genes. Mm-hmm.
  • fast_forward00:22:02 - So, but how long is the longest one you would find of all these 1,200 genes?
  • fast_forward00:22:08 - They're all very similar in length. Approximately 1,000 nucleotides of about 330 amino acids.
  • fast_forward00:22:14 - They have quite a short N-terminus and quite a short C-terminus.
  • fast_forward00:22:17 - So, a big part of the protein is presumably in the membrane. Okay, good.
  • fast_forward00:22:23 - So, as a family, as a total family, they're fairly uniform. So now we have this
  • fast_forward00:22:28 - strange phenomenon that some seem tightly clustered on the genome and others
  • fast_forward00:22:33 - are sort of singular, sitting out there somewhere in isolation.
  • fast_forward00:22:38 - So what could be the consequence of grouping all these genes together?
  • fast_forward00:22:43 - What could be a possible advantage in transcription or reproduction of a genome?
  • fast_forward00:22:50 - Is there anything you can say about that? Well, typically gene families are clustered.
  • fast_forward00:22:55 - That's the case in general genes that are related for which then multiple copies
  • fast_forward00:23:01 - typically come in a cluster they could have evolved by unequal crossing over
  • fast_forward00:23:07 - so by mistakes actually during meiosis we have an extra copy of the gene duplicated
  • fast_forward00:23:12 - created which then can then mutate away,
  • fast_forward00:23:17 - whether or not that creates an advantage is another issue but obviously having
  • fast_forward00:23:21 - genes with the same function clustered.
  • fast_forward00:23:26 - Opens the opportunity for local control of the cluster of regulatory elements that.
  • fast_forward00:23:32 - Decide or help decide within that cluster which of the genes is turned on right but now so in,
  • fast_forward00:23:40 - terms of the control this might already make a difference right so for olfaction
  • fast_forward00:23:43 - do you have in mind that there is such a variability and also these control
  • fast_forward00:23:48 - signals that for instance you want
  • fast_forward00:23:50 - to have one control signal to switch on a whole group of receptor genes while
  • fast_forward00:23:55 - you want to have a tighter control over another one.
  • fast_forward00:23:58 - This might be also, let's say, differential for, let's say, the different subsystems
  • fast_forward00:24:04 - that you would find in the epithelium, right?
  • fast_forward00:24:06 - Where maybe one, you might have very coarse control over the gene expression
  • fast_forward00:24:10 - and other regions, you want a very tight control over gene expression.
  • fast_forward00:24:13 - So do you mean in terms of numbers of cells that express it? For instance, yeah.
  • fast_forward00:24:17 - Yeah. Yeah, that's something that is, I think, underappreciated and also not
  • fast_forward00:24:21 - so easy to quantify is that you find often in papers also that each receptor
  • fast_forward00:24:28 - gene is expressed in one out of a thousand cells. Actually, it should be one in 1,200.
  • fast_forward00:24:31 - Even that is not true. There is two orders of magnitude of difference.
  • fast_forward00:24:36 - The champion is more 28. Right.
  • fast_forward00:24:38 - For which, as far as I know, still no ligand has been found,
  • fast_forward00:24:42 - and that's expressed in about 100,000 cells in a mouse. It's truly, again, a rule breaker.
  • fast_forward00:24:46 - There are other genes that are expressed in just a few hundred cells.
  • fast_forward00:24:50 - So the probability of expression, which we mean operationally,
  • fast_forward00:24:54 - the frequency or the number of cells in a given mouse that expresses a given
  • fast_forward00:24:57 - gene varies over two orders of magnitude.
  • fast_forward00:25:01 - Is that evolutionarily determined? Are the genes that are expressed in many
  • fast_forward00:25:06 - cells, are they more important? Therefore, there need to be more cells of them. I don't know.
  • fast_forward00:25:10 - But it's, of course, interesting for us in the long run to look at it experimentally.
  • fast_forward00:25:14 - What in the promoter, just upstream of the coding region, affects this probability. Right, exactly.
  • fast_forward00:25:22 - Because another angle on this could also be because you're saying,
  • fast_forward00:25:25 - well, from the perspective of olfaction, it looks haphazard, right? Like arbitrary.
  • fast_forward00:25:30 - But you could also argue, well, look, but it's not impossible that some of these
  • fast_forward00:25:35 - genes are expressed in other parts of the body of playing a different role.
  • fast_forward00:25:40 - And that's for that you want to have certain kinds of control sequences at work.
  • fast_forward00:25:45 - So that's an emerging story. There has been, ever since the beginning of the
  • fast_forward00:25:51 - auto receptor gene saga,
  • fast_forward00:25:52 - there's been papers about who are genes expressed in the testis,
  • fast_forward00:25:56 - particularly in sperm, but many genes are expressed there perhaps less interesting.
  • fast_forward00:26:00 - There has been some evidence that they may be involved in perhaps chemotaxis
  • fast_forward00:26:05 - of sperm, swimming towards the egg and so on, but that's all quite limited.
  • fast_forward00:26:10 - So apart from that, there are isolated cases of genes, odom receptor genes,
  • fast_forward00:26:15 - that are expressed outside the nose.
  • fast_forward00:26:17 - If they are also expressed in the nose and olfactory neurons,
  • fast_forward00:26:19 - they would qualify as an odom receptor, not with the O from odorant or olfaction.
  • fast_forward00:26:24 - If they're not expressed in the nose and only outside the nose,
  • fast_forward00:26:27 - then you cannot even call them a nodal receptor. They have just been hijacked
  • fast_forward00:26:31 - or used in evolution for something else.
  • fast_forward00:26:35 - Now, the one that comes to mind is this Moritine-2.
  • fast_forward00:26:39 - It has several other names, which is really a strange gene. It is expressed clearly in the nose.
  • fast_forward00:26:46 - There's ligands for it and so on and so on. They have glomeli.
  • fast_forward00:26:49 - It's also expressed in a small number of cells in kidneys.
  • fast_forward00:26:52 - Kidneys and mice with a knockout in that gene have a problem with regulating
  • fast_forward00:26:57 - their blood pressure, which is something you would never, never have thought to even look for.
  • fast_forward00:27:03 - So it's something that needs to be looked at further.
  • fast_forward00:27:07 - And indeed, in those cells in the kidney, the mechanisms that control the expression
  • fast_forward00:27:12 - are probably different from those in the nose. They're not random.
  • fast_forward00:27:16 - They're probably much more regulated and so on. So perhaps they have a different promoter.
  • fast_forward00:27:21 - The same coding region can have two promoters, a few kilobases apart,
  • fast_forward00:27:25 - and one promoter is used in the nose and the other one is used outside the nose.
  • fast_forward00:27:28 - That would be one simple solution.
  • fast_forward00:27:30 - Would you be willing to reconsider the notion of haphazard?
  • fast_forward00:27:35 - That maybe it looks haphazard because not all the constraints have been taken
  • fast_forward00:27:37 - into account, but if you would take on board the idea that these same receptors
  • fast_forward00:27:42 - are also expressed in other parts of the body, at other points during development,
  • fast_forward00:27:46 - for other functions possibly, Possibly that this requires other levels of control
  • fast_forward00:27:52 - that then require this kind of structuring of the genome. Would you find that
  • fast_forward00:27:56 - a reasonable alternative interpretation?
  • fast_forward00:27:58 - So I referred with haphazard to the organization in the genome,
  • fast_forward00:28:02 - not the expression per se. Okay.
  • fast_forward00:28:04 - I don't think there is rampant expression of auto-receptor genes outside the
  • fast_forward00:28:08 - nose, even in certain parts of development.
  • fast_forward00:28:11 - Probably would have been found by now.
  • fast_forward00:28:14 - So it's not the case that every odontoreceptor gene is at some point expressed somewhere else.
  • fast_forward00:28:20 - On the other hand, many senior libraries, many express sequence stacks,
  • fast_forward00:28:25 - ESTs, many microarrays that keep popping up odontoreceptors all the time to
  • fast_forward00:28:30 - the annoyance often of these investigators because they don't know what to do with it.
  • fast_forward00:28:33 - So I think it's been ignored a bit or underestimated too long in our field,
  • fast_forward00:28:40 - this non-olfactory expression.
  • fast_forward00:28:43 - But on the other hand, there is no rampant expression of OR genes outside the
  • fast_forward00:28:46 - nose, I think, any time in development.
  • fast_forward00:28:49 - Even though there is plenty of chemical sensing going on in other parts in the body.
  • fast_forward00:28:54 - Yeah, so these O-receptors in the end don't have a very unique receptor structure.
  • fast_forward00:29:00 - They're G-protein copper receptors, seven transmembrane proteins,
  • fast_forward00:29:02 - so they're amino acid sequence snakes seven times through the membrane.
  • fast_forward00:29:06 - Mean there's many other gpcrs in fact many of the drugs you buy in a pharmacy
  • fast_forward00:29:12 - with a prescription are designed against gpcrs agonist or antagonist and what
  • fast_forward00:29:18 - have you so from that point of view it's not surprising that some of these so-called
  • fast_forward00:29:23 - or genes so because they have an or like sequence,
  • fast_forward00:29:27 - are simply used by other cells to detect small molecules as their colleagues the non-or gpcrs,
  • fast_forward00:29:34 - non-olfactory GPCRs as they do, right?
  • fast_forward00:29:36 - Exactly right. That's indeed how pharmacologists talk now about their GPCR genes.
  • fast_forward00:29:41 - They call them non-olfactory GPCRs, by the way.
  • fast_forward00:29:44 - That's their GPCRs minus the 1200 that we work on.
  • fast_forward00:29:48 - But it might then also imply that the direction the field is taking is also,
  • fast_forward00:29:53 - well, maybe what we used to call an olfactory receptor gene,
  • fast_forward00:29:58 - is actually part of a larger family of, let's say, ligand-bound receptors Receptors.
  • fast_forward00:30:04 - And by accident, a subfamily of these have a specialization that they express
  • fast_forward00:30:08 - in the epithelium. Is that not where things are going now?
  • fast_forward00:30:13 - Well, I mean, this was known from the beginning. There are G-protein coupled
  • fast_forward00:30:16 - receptors, and that's an ancient motif for many receptor types.
  • fast_forward00:30:21 - So in that vein, 10 years ago, we reported this rather surprising finding.
  • fast_forward00:30:28 - We did what we call the receptor swap.
  • fast_forward00:30:29 - We replaced the coding region of a northern receptor, M71, for which we had ligands.
  • fast_forward00:30:35 - Acetophenolbenzaldehyde, we replace it with the beta-2 adrenergic receptor,
  • fast_forward00:30:38 - and that's the most, the best characterized GPCR, also I think the first one
  • fast_forward00:30:43 - to be cloned 20 years ago.
  • fast_forward00:30:45 - And lo and behold, the neuron that expressed now the beta-2 adrenergic receptor
  • fast_forward00:30:50 - from the M71 locus now form a new glomalus at a location quite far from the M71 glomali.
  • fast_forward00:30:58 - We also know that these neurons respond to beta-2AR agonists,
  • fast_forward00:31:03 - isoproterol, in a dose-dependent fashion.
  • fast_forward00:31:08 - So if you didn't know that in these mice the beta-2AR was expressed from this
  • fast_forward00:31:13 - M71 locus, and I would show you all these images, you would not be able to tell the difference.
  • fast_forward00:31:18 - So that's another, of course, artificial experimental evidence that perhaps
  • fast_forward00:31:24 - there is nothing so special about organ receptors in these regards.
  • fast_forward00:31:28 - Now, there are a lot of GPCRs expressed in the nose, in olfactory neurons.
  • fast_forward00:31:32 - That's another thing that's been neglected, perhaps a bit on purpose by our field.
  • fast_forward00:31:38 - Dozens and dozens of GPCRs are expressed. It's come up in several screens already, in several papers.
  • fast_forward00:31:43 - Some are expressed in olfactory neurons, like the dopamine type 2 receptor,
  • fast_forward00:31:47 - very frequent used for drugs against schizophrenia.
  • fast_forward00:31:52 - All olfactory neurons, all mature neurons express the dopamine type 2 receptor, which is also a GPCR.
  • fast_forward00:31:57 - And so why is the OR, can it instruct axons to form glomeruli and not other
  • fast_forward00:32:04 - GPCRs that are expressed there anyway is another issue for further research.
  • fast_forward00:32:08 - Perhaps it's a question of the timing of expression, the level,
  • fast_forward00:32:12 - or are they expressed at extremely high levels or something else we are missing. Right, exactly.
  • fast_forward00:32:19 - So now we have a bit of an idea of this sorting problem you've seen in the olfactory bulb.
  • fast_forward00:32:26 - Bulb, that means you have a huge population, a million plus receptor neurons
  • fast_forward00:32:32 - sitting there, sending their projections into the olfactory bulb,
  • fast_forward00:32:36 - initially in a rather disorganized way.
  • fast_forward00:32:41 - And then if you're by some sort of magic, it comes out sorted at the other end
  • fast_forward00:32:46 - and all these processes terminate in their preferred glomerulus.
  • fast_forward00:32:54 - And in some sense, we looked at the different aspects of that sorting story
  • fast_forward00:32:57 - that could play a role. And we see actually all of them are rather problematic.
  • fast_forward00:33:01 - Like the idea of zoning is problematic.
  • fast_forward00:33:06 - So what are the alternative interpretations of this? What would be an alternative
  • fast_forward00:33:10 - view of how this sorting process could work?
  • fast_forward00:33:13 - So a model that we proposed a while ago, and we talked about this this morning,
  • fast_forward00:33:17 - is that of homotypic and homophilic interactions.
  • fast_forward00:33:21 - Interactions, that the auto-receptor protein, which we know is expressed very
  • fast_forward00:33:25 - highly along the axons all the way to the end,
  • fast_forward00:33:27 - that auto-receptor proteins of the same type, so homotypic, interact with each
  • fast_forward00:33:32 - other, homophilic, and that would either by itself create some kind of a specific adhesion or perhaps,
  • fast_forward00:33:39 - combined with a signal transduction event in the axons, that would cause these
  • fast_forward00:33:44 - axons to fasciculate, to form fascicles, to form bundles,
  • fast_forward00:33:47 - which is probably a step that leads towards the formation of a glomerulus,
  • fast_forward00:33:50 - axons that stick together and at some point perhaps even stop growing and form
  • fast_forward00:33:56 - their glomerulus. That is one model that we favor.
  • fast_forward00:33:59 - It would be a parsimonious model evolutionarily because if a new-ordered receptor
  • fast_forward00:34:03 - is created in evolution with an amino acid sequence that's sufficiently different.
  • fast_forward00:34:08 - Then these homophilic interactions are different now, and now you would have
  • fast_forward00:34:13 - a new identity and a new glomerulus created without anything else.
  • fast_forward00:34:17 - But wait, I'm not sure if that's the whole story then, because you would still
  • fast_forward00:34:21 - might have to generate an overabundance of these processes to have this sorting
  • fast_forward00:34:27 - process, self-organizing sorting to work out,
  • fast_forward00:34:30 - because you lack specificity now. You mean neurons and axons?
  • fast_forward00:34:33 - That's right. I mean, these receptor neurons that throw out these processes,
  • fast_forward00:34:36 - if they have these attractive and repellent interactions, you might want to
  • fast_forward00:34:41 - throw out quite a large number of them in the hope that a small subset will
  • fast_forward00:34:46 - actually reach the target or not.
  • fast_forward00:34:47 - So that's the issue of selection and, again, something that is, I think, ignored or….
  • fast_forward00:34:54 - Overlooked in the field is how many, what's the success rate of the whole process?
  • fast_forward00:35:00 - To put it in a simple way, of every hundred neurons in the epithelium that are born and project an axon,
  • fast_forward00:35:06 - that gets to the bulb, how many of those send their axon to the correct place
  • fast_forward00:35:13 - in the sense that it innervates a correct glomalus and it survives, right?
  • fast_forward00:35:16 - That's a very simple question. What is the success rate? And we don't know that.
  • fast_forward00:35:20 - We intuitively think that it could be very high, Perhaps as close as 100%,
  • fast_forward00:35:24 - but there's no such thing in biology.
  • fast_forward00:35:26 - If it's quite low, though, let's say it's less than 50%, then there is an opportunity
  • fast_forward00:35:30 - for selection, for negative selection, if you wish, for weeding out processes,
  • fast_forward00:35:35 - as you call them, axons that are completely off the wall and project to the
  • fast_forward00:35:39 - wrong part of the bulb and are hopelessly lost.
  • fast_forward00:35:44 - Or axons that enter a glomalus that is very close, but not exactly of the same
  • fast_forward00:35:48 - type, right? And they could, over a period of hours of days, being eliminated.
  • fast_forward00:35:53 - So there is no method to look at the success rate.
  • fast_forward00:35:57 - And also when people look at these images, I don't think they think about that,
  • fast_forward00:36:02 - that some axons could not make it.
  • fast_forward00:36:05 - It's just not, in that sense, these pictures are somewhat misleading perhaps.
  • fast_forward00:36:09 - They give you the end product, the final results, but it doesn't tell you about
  • fast_forward00:36:13 - the mechanism necessarily.
  • fast_forward00:36:15 - Possibly in different stages of development, you might see signatures of that process at work.
  • fast_forward00:36:20 - So in general, the brain has an exuberance, as it's often called.
  • fast_forward00:36:25 - More neurons produced than necessary, more axonal processes produced than necessary.
  • fast_forward00:36:30 - That's not the case, by the way, with olfactory axons. They never have multiple
  • fast_forward00:36:34 - processes, just one per, at least that's a dogma, one per neuron.
  • fast_forward00:36:40 - But again, I would not be surprised at all if someone finds out that the success rate is very small,
  • fast_forward00:36:45 - very low, that there is an exuberance of neurons being produced,
  • fast_forward00:36:49 - that many of them never make it, and they get removed so quickly from the system
  • fast_forward00:36:55 - within hours of days, and everything is asynchronously developing anyway,
  • fast_forward00:36:59 - that we would just be missing them.
  • fast_forward00:37:00 - But if you would look specifically for them, and really with very specific methods,
  • fast_forward00:37:04 - perhaps you could find them. There is a lot of cell death going on in the olfactory epithelium.
  • fast_forward00:37:08 - You can use any marker you wish for apoptosis, even in adult mice.
  • fast_forward00:37:13 - Plenty and plenty of cells dying, as it decays in every epithelium.
  • fast_forward00:37:17 - So at least there is a basis there for cell death playing a role in scalloping,
  • fast_forward00:37:23 - if you wish, these projections.
  • fast_forward00:37:26 - Right. So now, another step, because there's a lot of work behind these observations
  • fast_forward00:37:33 - that you're sharing now with me.
  • fast_forward00:37:36 - But then as one step in that whole process, you also decided to clone a mouse.
  • fast_forward00:37:42 - So how did that help you? How did that help you in understanding this system?
  • fast_forward00:37:45 - So that was for the gene regulation issue. issue, a very popular idea back from
  • fast_forward00:37:51 - the early days from 1991,
  • fast_forward00:37:52 - the Bercouzel paper, was that the way an olfactory neuron expresses one gene
  • fast_forward00:37:59 - stably and irreversibly at high levels is by some kind of genetic alteration in the genome.
  • fast_forward00:38:07 - That's the case with lymphocytes.
  • fast_forward00:38:09 - B cells make one antibody with a heavy and a light chain, and T-cell receptors,
  • fast_forward00:38:14 - the Let's go for beta t's as if one alpha and one beta chain.
  • fast_forward00:38:19 - And they do so by irreversibly changing their genetic material.
  • fast_forward00:38:23 - There are pieces of DNA that are lost, obviously irreversible,
  • fast_forward00:38:27 - and in some cases inverted, that underlie this.
  • fast_forward00:38:34 - But surely the problem of cell faith is broader than that, right?
  • fast_forward00:38:38 - It's not only for the lymphocytes. I mean, any cell at some point in development
  • fast_forward00:38:42 - is committed to become of a certain identity, like a skin cell or a liver cell or a heart cell, etc.
  • fast_forward00:38:50 - So why do you highlight the lymphocytes in this case?
  • fast_forward00:38:53 - Because they have so many similarities with olfactory neurons.
  • fast_forward00:38:56 - They have a large number of genes at their disposal, and each neuron or each
  • fast_forward00:39:02 - lymphocyte expresses only one of them. I mean, it's not only me who makes this analogy.
  • fast_forward00:39:06 - So this was pervasive in our thinking. I must say that many of the people working
  • fast_forward00:39:11 - in olfaction, at least in the early days, were actually ex-immunologists,
  • fast_forward00:39:14 - had a PhD or a postdoc in immunology, so maybe they were thinking along these lines.
  • fast_forward00:39:19 - But don't you agree that today, we might as well say, well, the analog might
  • fast_forward00:39:24 - as well have been a skin cell?
  • fast_forward00:39:26 - Well, but skin cells don't have to acknowledge a large number of genes of the
  • fast_forward00:39:30 - same family at the disposal of which they have to pick one for expression.
  • fast_forward00:39:33 - That's just not the case.
  • fast_forward00:39:35 - But in the end, they have to pick a subset of all possible genes to express.
  • fast_forward00:39:38 - So, to go back even longer in history,
  • fast_forward00:39:43 - when these gene rearrangements in B lymphocytes were discovered in the 70s by
  • fast_forward00:39:48 - Susumu Tonagawa and others, there was the idea that other aspects of differentiation
  • fast_forward00:39:53 - and development would also be regulated by irreversible gene rearrangements.
  • fast_forward00:39:57 - But the cloning experiments of John Gurdon and others were a strong argument
  • fast_forward00:40:01 - against that because you could produce normal or fairly normal animals from differentiated cells.
  • fast_forward00:40:07 - And if these cells had shattered pieces of DNA irreversibly,
  • fast_forward00:40:11 - then they would not have been able to do that.
  • fast_forward00:40:13 - But indeed, there was a period of time, 30, 40 years ago, where it was thought
  • fast_forward00:40:18 - that many differentiation processes would be….
  • fast_forward00:40:25 - Governed by irreversible changes. And we don't think that's the case anymore.
  • fast_forward00:40:29 - So it's basically like, look, your faith is set by some switch.
  • fast_forward00:40:32 - And now basically, in some sense, you irreversibly wipe out a bit of your genome.
  • fast_forward00:40:36 - So from then on, that's it.
  • fast_forward00:40:38 - And also the discovery of the possibility of making induced pluripotent stem
  • fast_forward00:40:43 - cells, the iPS cells, was another argument against that.
  • fast_forward00:40:45 - You can take a differentiated cell and sort of reconvert it into an embryonic
  • fast_forward00:40:50 - stem cell or like cell that can go any way.
  • fast_forward00:40:53 - Way that's also an argument against that irreversible changes
  • fast_forward00:40:56 - of any type actually not necessarily genetic but then
  • fast_forward00:40:59 - how did this this cloned mouse that that you build
  • fast_forward00:41:02 - was quite a tour in itself to to build it
  • fast_forward00:41:05 - help you to understand this this determination and stabilization of of the cell
  • fast_forward00:41:13 - faith in the olfactory system so if if neurons that express a given receptor
  • fast_forward00:41:16 - let's take our favorite receptor again m71 if neurons expressing that receptor
  • fast_forward00:41:20 - have made an irreversible genetic alteration to do so,
  • fast_forward00:41:25 - then a mouse cloned from the nucleus of such a neuron would be monoclonal, essentially.
  • fast_forward00:41:31 - All the neurons, or the majority of them, would express M71.
  • fast_forward00:41:35 - That was the prediction. If these changes are irreversible, if they're somehow
  • fast_forward00:41:39 - reversible during the cloning procedure, which not many people have thought
  • fast_forward00:41:42 - about, but I sometimes think about this, if they would have been reversible
  • fast_forward00:41:45 - during the cloning procedure, then we draw the wrong conclusion.
  • fast_forward00:41:49 - So it was essentially a negative result. We got, let's say, a normal mouse out
  • fast_forward00:41:53 - of the nucleus of a neuron expressing M71.
  • fast_forward00:41:56 - These neurons expressed not necessarily M71, but other own receptor genes.
  • fast_forward00:42:00 - Hence, there were no irreversible changes. If you clone a mouse with a lymphocyte,
  • fast_forward00:42:05 - Rudy Yenish has done that around the same time, then you really get a monoclonal
  • fast_forward00:42:09 - mouse. These are spectacular phenotypes.
  • fast_forward00:42:11 - All the B lymphocytes in that mouse make just the same antibody.
  • fast_forward00:42:14 - Right. But how come that cloned mouse is even viable?
  • fast_forward00:42:18 - It might have been possible. It would not even have been a viable cell. Thank you for watching.
  • fast_forward00:42:22 - Yeah, I think our paper and that of Rudy Inej and Richard Axel was the first
  • fast_forward00:42:26 - to clone with post-mitotic mature neurons.
  • fast_forward00:42:29 - That was not necessarily given.
  • fast_forward00:42:33 - The success rate of cloning by nuclear transfer is quite low,
  • fast_forward00:42:36 - I must say. It's in the single-digit percentages.
  • fast_forward00:42:39 - Some of the data are a bit massaged in these tables, but it's never more than 10%.
  • fast_forward00:42:46 - And so it could still be that the other 90% die for fundamental biological reasons.
  • fast_forward00:42:51 - We will only know when we're able to get the frequency higher.
  • fast_forward00:42:55 - Right. Okay, but so now that you know that there might still be an irreversible
  • fast_forward00:43:00 - change, but it doesn't translate if you clone from that cell a whole new mouse, right?
  • fast_forward00:43:07 - So what now is your alternative interpretation?
  • fast_forward00:43:12 - Well, if it's not genetic, it must be epigenetic. That's the more modern,
  • fast_forward00:43:17 - perhaps fashionable way of looking at that.
  • fast_forward00:43:19 - But wait, that could also be, imagine I have another cell that's expressing
  • fast_forward00:43:25 - other genes that is controlling the expression of the genes in this receptor neuron.
  • fast_forward00:43:31 - So now you clone only from this receptor neuron, but this controlling unit is gone.
  • fast_forward00:43:36 - Now the control signal is gone. So this receptor neuron again is expressing its full genome.
  • fast_forward00:43:43 - Yes, but at least that gene or receptor gene that was expressed in the original
  • fast_forward00:43:47 - cell, that gene was not irreversibly altered.
  • fast_forward00:43:50 - That's right, exactly. That's what we said, nothing more. Okay,
  • fast_forward00:43:53 - so not irreversibly changed just within that one cell.
  • fast_forward00:43:59 - So you do consider the possibility that within the organism there might be additional
  • fast_forward00:44:03 - control signals that regulate that.
  • fast_forward00:44:06 - Well, what I was worried then and still a bit now is that experimentally during
  • fast_forward00:44:10 - the process of nuclear transfer, which is a complex procedure,
  • fast_forward00:44:13 - which we don't understand, that change that was present in the M71 locus was somehow reverted.
  • fast_forward00:44:20 - If you have a thing that flips back and forth, it just flips back, right?
  • fast_forward00:44:25 - And then we don't see it anymore, but that doesn't mean there was no change
  • fast_forward00:44:28 - before that. Sure, absolutely.
  • fast_forward00:44:30 - Because maybe the whole confirmation of the DNA was such that it was put into
  • fast_forward00:44:43 - a metastable state that prevented further transcription.
  • fast_forward00:44:48 - But that by, let's say, perturbing the system again in the cloning procedure,
  • fast_forward00:44:53 - it could sort of flip back into a state that could express itself.
  • fast_forward00:44:56 - I can see that. But so now if you say epigenetic, that's of course a little
  • fast_forward00:45:01 - bit tricky because at this point, epigenetic just means, well,
  • fast_forward00:45:05 - some factor that's not genetic, right?
  • fast_forward00:45:08 - But that basically could be from the whole of the universe to the diet to other
  • fast_forward00:45:13 - cells, developmental trajectories, whatever.
  • fast_forward00:45:16 - So if you say epigenetic, what would you have exactly in mind?
  • fast_forward00:45:20 - Well, that's not what I say, but that's what other people say.
  • fast_forward00:45:23 - And indeed, it's very broad. I mean, in some ways it doesn't say very much because
  • fast_forward00:45:28 - it only says it's not genetic.
  • fast_forward00:45:29 - That's exactly right. That we sort of knew there is no genetic change. The DNA isn't changed.
  • fast_forward00:45:34 - We think it's not changed in the organ receptor gene locus. But I would…,
  • fast_forward00:45:40 - I think that at this point in time, we really don't have a good view of how,
  • fast_forward00:45:44 - I mean, I'm the first to admit that, how a neuron expresses one allele of one
  • fast_forward00:45:49 - gene at high levels. We just don't understand that.
  • fast_forward00:45:52 - It sort of have clues of why it could be a small number of genes,
  • fast_forward00:45:55 - but why it's just one, that is very difficult for anyone to think on or show experimentally.
  • fast_forward00:46:03 - Okay, but then what's the next step there to solve this?
  • fast_forward00:46:07 - Probably development of new technologies miniaturization we have to work with
  • fast_forward00:46:12 - single cells uh one can sort cells expressing the same receptor using a cell
  • fast_forward00:46:16 - sorter with gfp but even then you look at the population of cells so we have
  • fast_forward00:46:20 - to look at single cell and that's all coming uh coming along now there's rna
  • fast_forward00:46:25 - seek is now being developed for single cells,
  • fast_forward00:46:28 - um so i think it's a as usual new new technologies will um give us new ways
  • fast_forward00:46:33 - of looking at all problems.
  • fast_forward00:46:36 - But without a clear hypothesis, the technology might also lead you astray.
  • fast_forward00:46:43 - Yeah. Or hypotheses, multiple hypotheses.
  • fast_forward00:46:48 - You want to ask, basically, neurons that express receptor A,
  • fast_forward00:46:52 - how are they different from neurons expressing receptor B?
  • fast_forward00:46:54 - And let's say that A and B are two similar genes in the same cluster, as close as you can get.
  • fast_forward00:46:59 - What is different between those cells? That would be a very simple approach approach, right?
  • fast_forward00:47:05 - And you would take anything you find differentially expressed or differentially
  • fast_forward00:47:08 - methylated or hypomethylated because one of them could be causative.
  • fast_forward00:47:12 - Others could be the consequence of neurons expressing receptor A versus B,
  • fast_forward00:47:16 - but other changes or differences could be explained or help explain why it's
  • fast_forward00:47:21 - receptor A in one versus the other.
  • fast_forward00:47:24 - But I think in the long run, we have to look at single cells, but it's possible now.
  • fast_forward00:47:29 - There is really a big, big progress in single cell studies.
  • fast_forward00:47:33 - But you think to do that effectively, you do need new technologies.
  • fast_forward00:47:38 - Yeah, but that's always been the case, no? Well, it depends.
  • fast_forward00:47:42 - I mean, given the question you posed, you might have to develop a specific technology,
  • fast_forward00:47:47 - but you don't have to wait for things to show up at the horizon.
  • fast_forward00:47:50 - But I mean, in your research, you're very technology heavy in your approach, right?
  • fast_forward00:47:55 - So one thing you described is this nanostring system that you have been using
  • fast_forward00:48:02 - to identify, and also other aspects of this whole regulatory system
  • fast_forward00:48:07 - within the cell for gene expression around this notion of the P element.
  • fast_forward00:48:13 - So what has been the insight there with respect to this regulation question?
  • fast_forward00:48:19 - So we have adopted nanostring because there was probably still is no good method
  • fast_forward00:48:23 - to look at all these organ receptor genes at the same time in the same sample.
  • fast_forward00:48:28 - Typical way of looking at it is by qPCR, quantitative PCR, but that's really
  • fast_forward00:48:33 - asking a lot doing that for hundreds of genes from one sample, a lot of pipetting.
  • fast_forward00:48:37 - There's microarrays. I think their specificity is somewhat questionable, to say it nicely.
  • fast_forward00:48:45 - Often these probes are from the 3-prime non-translated region,
  • fast_forward00:48:48 - which is computationally determined, so not experimentally validated,
  • fast_forward00:48:53 - because it's difficult to do so.
  • fast_forward00:48:55 - So we have chosen for this nanostring, with which, since we chose only to make
  • fast_forward00:49:00 - probes against all receptor coding regions, we could only look at at half the genes.
  • fast_forward00:49:04 - So it's for us an assay. We can really look at close to 600 genes have the repertoire
  • fast_forward00:49:09 - in one sample of RNA of about a microgram without any kind of amplification.
  • fast_forward00:49:14 - It's an assay. I mean, by itself it doesn't show anything. And with that assay,
  • fast_forward00:49:18 - we showed that mice that lacked at a mutation in this 317 base pair element,
  • fast_forward00:49:25 - which we call the P element.
  • fast_forward00:49:27 - That there are 10 genes differentially expressed, nine are downregulated,
  • fast_forward00:49:32 - there is less RNA in the mutant mouse versus the wild type, and one is slightly upregulated.
  • fast_forward00:49:36 - And they're all within about 200 kilobase from the so-called P elements.
  • fast_forward00:49:41 - So that shows that that element somehow regulates the expression of OR genes
  • fast_forward00:49:48 - in the cluster at the organism level.
  • fast_forward00:49:51 - And you believe that this might be a key step in determining the faith of these receptor neurons?
  • fast_forward00:49:57 - Yeah, one of them. It's not the only one. One, there is, we also said that clearly in our paper,
  • fast_forward00:50:04 - that it's of course not by itself explaining how an olfactory neuron expresses
  • fast_forward00:50:08 - one allele of one gene, but it's one of the several layers,
  • fast_forward00:50:14 - Possibly hierarchical layers of control mechanisms that altogether ensure that
  • fast_forward00:50:19 - the large majority, if not all, mature neurons expressing one allele of one gene.
  • fast_forward00:50:23 - But by itself, it cannot be responsible for that.
  • fast_forward00:50:27 - So how complex do you think is this control hierarchy? Yeah.
  • fast_forward00:50:31 - Again, we will answer that when we know the whole hierarchy.
  • fast_forward00:50:36 - It's not going to be solved immediately. I still have many years to go before my pension.
  • fast_forward00:50:43 - And it has multiple aspects to it, multiple levels. There is also cellular selection.
  • fast_forward00:50:48 - I believe that occasionally neurons are produced that make two receptors or maybe even more.
  • fast_forward00:50:53 - And perhaps they are lost somehow, lost in the system by negative selection
  • fast_forward00:50:57 - and that you can look for that too.
  • fast_forward00:50:59 - It would be another control mechanism, right? You produce, you eliminate the cells you don't want.
  • fast_forward00:51:04 - So it's not going to be solved anytime soon. The journals, of course,
  • fast_forward00:51:08 - like us to claim in our papers in top journals in the title and so on that this
  • fast_forward00:51:11 - is now the final breakthrough.
  • fast_forward00:51:15 - But how many layers of control do you expect? Is it like single digits? 17.
  • fast_forward00:51:23 - No, multiple. And at a different genomic level,
  • fast_forward00:51:27 - positioning of clusters for expression, there's some evidence for that recently
  • fast_forward00:51:33 - with nuclear aggregation that the gene that's expressed is in a different position
  • fast_forward00:51:40 - of the nucleus than the other genes.
  • fast_forward00:51:42 - That could be one of the mechanisms, but it cannot explain why it's only one, right? Right.
  • fast_forward00:51:46 - That's always the problem. Why is it just one?
  • fast_forward00:51:51 - But it's not, I mean, I think a satisfying level of insight,
  • fast_forward00:51:56 - the question is how much, when are you satisfied with saying, I've understood this.
  • fast_forward00:52:00 - A satisfying level of insight is feasible, conceivable, the next decade or two, but not sooner.
  • fast_forward00:52:08 - And we hope, of course, that it has some general relevance, that we have not
  • fast_forward00:52:15 - just explained how this particular weird family of genes is controlled,
  • fast_forward00:52:19 - but that it has some insights that are more generally relevant to biology.
  • fast_forward00:52:22 - Exactly right, because we started the conversation with how you use the olfactory
  • fast_forward00:52:27 - system as a model to study development.
  • fast_forward00:52:29 - Yeah. Right? But now, in some sense, we ended up dealing with a question that
  • fast_forward00:52:33 - seems rather specific for the olfactory system, which is, how do I get such
  • fast_forward00:52:37 - a precise expression of a single gene in a single neuron?
  • fast_forward00:52:40 - So, what's this telling us in the end about development?
  • fast_forward00:52:46 - Well, again, we will say this at the end, whether it was worth it and whether
  • fast_forward00:52:50 - it was. I want to know it now.
  • fast_forward00:52:52 - I was once a few years ago with Michael Brown in Beijing, coincidentally at
  • fast_forward00:52:58 - the same time at an institute, and the graduate students all asked him,
  • fast_forward00:53:01 - you know, how do I find an interesting biological problem, right,
  • fast_forward00:53:04 - that gives me a Nobel Prize like he got?
  • fast_forward00:53:07 - And he had a very good answer. They said you can pick almost any problem to
  • fast_forward00:53:10 - start with, and you have to dig deeper and deeper and deeper.
  • fast_forward00:53:12 - And when you hit very deeply, you will come to fundamental principles and mechanisms.
  • fast_forward00:53:17 - So the angle or the approach point at which you start to study biological phenomenon
  • fast_forward00:53:24 - or question perhaps matters less. It's how deep you go.
  • fast_forward00:53:27 - But I constantly try to remind myself that we are not working on just the sense of smell of a mouse.
  • fast_forward00:53:33 - You could just say that very honestly, that we hope that what we find and what
  • fast_forward00:53:39 - we uncover, that is more important than just olfactory system of the mouse.
  • fast_forward00:53:44 - That is a conscious leitmotif.
  • fast_forward00:53:47 - If you wish, it's a hope and a plan.
  • fast_forward00:53:54 - Right. So then to get to the finish line, two things.
  • fast_forward00:53:58 - So, actually, you told me that it's exactly 20 years ago that you got initiated
  • fast_forward00:54:04 - in this experimental study of olfaction, when you entered the lab of Axel as
  • fast_forward00:54:08 - a postdoc. Yeah. That's correct, yeah?
  • fast_forward00:54:12 - And so, now, given your experience in this field and also your many accomplishments,
  • fast_forward00:54:16 - what would be Peter's law that we should adhere to in trying to understand how the brain works?
  • fast_forward00:54:24 - Well, I don't like this question to begin with, how the brain works.
  • fast_forward00:54:27 - That is so overly ambitious and probably even nonsensical, you can only hope
  • fast_forward00:54:32 - at least with the present technologies to understand a bit of it, an aspect of it.
  • fast_forward00:54:38 - Perhaps one circuit, one small step, it's some kind of Flemish modesty perhaps,
  • fast_forward00:54:44 - that we, you know, perhaps with time and when we understand more smaller bits
  • fast_forward00:54:49 - of it, more deeper and fundamental, we can… But I definitely never said that
  • fast_forward00:54:54 - or never will claim that we want to figure out how the brain works.
  • fast_forward00:54:56 - That is such a big, big question so far away.
  • fast_forward00:54:59 - One has to maybe say this in grant proposals.
  • fast_forward00:55:02 - So Peter's law is to be modest.
  • fast_forward00:55:05 - And try to answer a problem that
  • fast_forward00:55:08 - is answerable at that time with the technologies available at that time.
  • fast_forward00:55:12 - But I think other people have said that too and realized that too.
  • fast_forward00:55:15 - It's solving the solvable or Peter Medawar said that, right? Okay.
  • fast_forward00:55:19 - But then the other hand is, so five years from now, I'm going to go to Frankfurt.
  • fast_forward00:55:25 - For it and I'm going to confront you with your predictions of today.
  • fast_forward00:55:29 - So what's the prediction that you're most passionate about today that you really
  • fast_forward00:55:35 - spent most of your time on that you really want to have tested five years from
  • fast_forward00:55:39 - now that I can confront you with then? Well the.
  • fast_forward00:55:44 - But that's, again, technologically is the single-cell studies,
  • fast_forward00:55:46 - I think. But I'm not the only one who's thinking that. But I want a prediction, Peter.
  • fast_forward00:55:50 - But that is coming so clearly there, and there's not much published,
  • fast_forward00:55:53 - and it will give us really new insights.
  • fast_forward00:55:56 - Because we typically, when we look at the biological phenomenon,
  • fast_forward00:56:00 - we look at the population of cells, and we are missing a lot.
  • fast_forward00:56:02 - For instance, to come back to our p-element, the p-element shows a reduction
  • fast_forward00:56:08 - in expression at the population level.
  • fast_forward00:56:11 - We all hope that we have a nice Gaussian curve, right, which is shifted.
  • fast_forward00:56:15 - But if the Gaussian curve is split into a bimodal curve, you would get exactly the same results.
  • fast_forward00:56:21 - And it's actually erroneous results because we're now misled.
  • fast_forward00:56:24 - So I really believe for our system and the things we are interested in,
  • fast_forward00:56:28 - we have to look at single cells. It is doable.
  • fast_forward00:56:31 - We are doing and other people will do it. And in five years,
  • fast_forward00:56:33 - I think there will be nice insights coming from that.
  • fast_forward00:56:36 - But that's a very modest prediction, right? So you're saying five years from
  • fast_forward00:56:39 - now, I'll be able to look into single cells in the olfactory system.
  • fast_forward00:56:44 - And see their genetic expression dynamics.
  • fast_forward00:56:47 - It's already possible, yes, with RNA-seq. Yeah, but if you can already do it,
  • fast_forward00:56:50 - it's not a very exciting prediction for five years from now.
  • fast_forward00:56:52 - Well, but also, do we get something interesting out of that?
  • fast_forward00:56:55 - Or will we be lost in details or in variability and so on?
  • fast_forward00:57:00 - I hope and I believe that there will be new insights coming out of it,
  • fast_forward00:57:03 - which we didn't even imagine.
  • fast_forward00:57:04 - But would you say five years from now, you have nailed this problem of the cell
  • fast_forward00:57:08 - phase stabilization? No, not at all.
  • fast_forward00:57:10 - No? No, I don't want to put any year on that.
  • fast_forward00:57:14 - That's really a longer project five year plans for the communists I know that,
  • fast_forward00:57:19 - that's why I'm asking so Peter Wombart and for grant organizations so Peter
  • fast_forward00:57:25 - Wombart, thank you very much for this conversation thanks Paul.
  • fast_forward00:57:28 - Music.
  • fast_forward00:57:34 - The CSN podcast was produced by the Convergent Science Network of Biometrics
  • fast_forward00:57:40 - and Biohybrid Systems A project funded by the European Sevens Research Framework Programme.
  • fast_forward00:57:48 - For more interviews, recorded lectures or upcoming conferences in the field
  • fast_forward00:57:53 - of biometrics and biohybrid systems, go to csnnetwork.eu.
  • fast_forward00:58:00 - Thank you.
  • fast_forward00:58:01 - Music.

Be the first to leave a comment

Leave a comment

Your email address will not be published. Required fields are marked *

Convergent PozitivLinia Logo

Exploring the convergence of neuroscience, robotics, and AI through conversations with leading researchers since 2010.

A project of the Convergent Science Network Foundation.

© CSN Podcasts. Developed by IMCreativeWEBC

0%

Login to enjoy full advantages

Please login or subscribe to continue.

Go Premium!

Enjoy the full advantage of the premium access.

Stop following

Unfollow Cancel

Cancel subscription

Are you sure you want to cancel your subscription? You will lose your Premium access and stored playlists.

Go back Confirm cancellation